The Addenbrooke’s Cognitive Examination–Revised in the differential diagnosis between Alzheimer's disease and behavioral variant frontotemporal dementia: A new model

Abstract

AbstractBackgroundThe increase in life expectancy implies the emergence of chronic‐degenerative and disabling conditions, such as cognitive impairment and dementia. Among the frequent disorders in clinical practice, the differential diagnosis between Alzheimer's disease (AD) and behavioral variant frontotemporal dementia (bvFTD) may be challenging, which justifies the improvement in cognitive tools to be used in clinical practice.MethodThe Addenbrooke’s Cognitive Examination‐Revised (ACE‐R) was administered to 102 patients diagnosed with mild dementia due to probable AD and 37 with mild probable bvFTD from two Brazilian research centers (Table 1). All individuals were submitted to the Mattis Dementia Rating Scale (DRS) and the ACE‐R. The performance of the patients was compared and analyzed. ROC curve analyses were conducted to assess the ACE‐R accuracy in the diagnosis of AD and bvFTD. Multivariate logistic regression analysis was used to develop a new model to refine the diagnostic accuracy of the ACE‐R.ResultMean total scores in the ACE‐R were 70.2 (standard deviation ‐ SD = 10.8) in patients with AD and 72.2 (SD = 11.1) in bvFTD. The VLOM ratio showed an area under the curve (AUC) of 0.816, with 87% sensitivity (Sens.) and 71% specificity (Spec.), 73% positive predictive value (PPV) and 86% negative predictive value (NPV) for the differential diagnosis between AD and bvFTD. The logistic regression method with cross‐validation showed that the relationship between Attention and Orientation, Fluency, Language and Age, as independent variables, shared the importance to the diagnostic differentiation between bvFTD and AD (p <0.05; Table 2). The proposed logarithm was superior in discriminating bvFTD and AD than the VLOM ratio, with AUC of 0.865 (Figure 1) with 78% Sens., 85% Spec., 65% PPV and 91% NPV (Table 3).ConclusionThe ACE‐R achieved very good diagnostic power to discriminate AD and bvFTD in the present sample. Furthermore, the final ROC curves showed the superiority of the model proposed in relation to the analysis of the subscales individually. Further analysis in larger samples, with biomarkers or pathological confirmation, are necessary to confirm these findings.